The present invention relates to a controller for an internal combustion engine executing spark-ignited combustion and pre-mixed compression-ignited combustion. More specifically, the invention relates to a technical matter intended for stabilization of compression-ignited combustion even when an engine or an actuator becomes a deterioration condition.
In an internal combustion engine which basically performs spark-ignited combustion, a method in which a part of the engine operation area adopts compression-ignited combustion system instead of spark-ignited combustion system is known. This method makes it possible to reduce NOx emissions and highly efficiently operate the engine. It is expected to optimally control the compression-ignited combustion so as to provide a stable compression-ignited combustion operation over a wide operation range.
According to the description of JP-A No. 82229/2001 or JP-A No. 108218/2003, it is effective for the compression-ignited combustion to control the combustion by means of a mass of internal EGR (fuel injection during a negative valve overlap period) using a variable valve mechanism for an inlet valve and outlet valve of an engine cylinder or by means of a multiple injection.
In the compression-ignited combustion, there is no physical ignition equipment such as spark plug and the piston compression allows a pre-mixed mixture to self-ignite. The control over ignition timing or the like requires accurate fuel injection control or valve timing control. When an actuator for the fuel injection system or the variable valve becomes deteriorated, the compression-ignited combustion is subject to large variations between cylinders or cycles. The stable compression-ignited combustion becomes unavailable.
There is a need for a solution of implementing the stable compression-ignited combustion even when the actuators become deteriorated. According to the technology described in JP-A No. 108218/2004, for example, in order to improve the compression-ignited combustion, the negative valve overlap period or the fuel injection quantity during the negative valve overlap period is changed in accordance with a peak value or a peak timing of a cylinder pressure in the compression-ignited combustion. In this manner, the pressure peak value or the pressure peak timing can be controlled to an appropriate value even when the actuator for the engine or the fuel injection system becomes deterirated.
The technology described in the patent document performs combustion control by using combustion control means such as the variable valve mechanism and the fuel injection quantity based on a combustion state recognized during the compression-ignited combustion. However, when the deterioration of the engine or its actuators become in remarkable advanced stage, such combustion control can be not enough good any more to the deterioration, as a result, temporary knocking or misfire may occur before or during the compression ignition control.
On the other hand, when the multiple injection controls stabilization of the compression-ignited combustion, the applicants experimentally confirmed that a low load area and a high load area must use one-time fuel injection equivalent to approximately a Minimunm Reliable fuel injection quantity (the minimum fuel injection quantity ensuring stable injection for the fuel injection system) of the fuel injection system. Even when normal injection pulse width correction is performed for a deteriorated fuel injection system, these operation areas may not be able to use a specified fuel injection quantity according to the minimum reliable injected quantity. The above-mentioned patent documents give no consideration to these problems and make no mention of a technique of detecting the fuel injection system deterioration or a specific technique of controlling the deterioration detection (recognition).